The sensitivity of the chi-square test (Chap. 11) and the odds ratio test (Chap. 12) for testing cross-tabs is limited, and not entirely accurate if the values in one or more cells is smaller than 5. The log likelihood ratio test is an adequate alternative with generally better sensitivity, and, so, it must be absolutely recommended.
CITATION STYLE
Cleophas, T. J., & Zwinderman, A. H. (2011). Log Likelihood Ratio Tests. In Statistical Analysis of Clinical Data on a Pocket Calculator (pp. 37–38). Springer Netherlands. https://doi.org/10.1007/978-94-007-1211-9_13
Mendeley helps you to discover research relevant for your work.